Because I am a nerd with too much time on his hands, I think about buying the rights to broadcast the Olympics in Australia and what that entails. Here’s some back of the envelope type hypothesising on the costs, expenses and possible profit if anyone wanted to buy the rights to the 2024 Olympics.
The 7 Network purchased broadcast rights off the IOC for between $150m-$170m, according to this SMH story. The real figures are not made public, so all we can go on is this story and other stories at say it was definitely lower than $200m. So let’s be generous and assume 7 paid $180m to be the sole broadcaster of the 2016, 2018 and 2020 Olympics in Australia.
To split this out into actual numbers for each Olympics, we can assume the winter Olympics in 2018 is the least valuable to 7, as Australian’s don’t give too much of a rats arse about snow sports. The two summer Olympics are differentiated by the time zone and what can be shown during prime time. Tokyo’s (2020 Olympics host city) time zone is much more conducive to prime time viewing than Rio, so I assume 7 values that a bit higher than the 2016 Olympics. With that in mind – 2016 Olympics: $65m, 2018 Olympics: $35m, 2020 Olympics: $80m.
This puts the value of the 2020 Olympics – an Olympics in our time zone, at around $80m.
We obviously can’t buy the rights to the 2020 Olympics – they’re already sold, but the 2024 Olympic rights are up for grabs. The current bid cities for 2024 are Rome, Paris, Budapest and Los Angeles. All of which are in relatively poor time zones for Australian viewers compared to an Asian Olympics, decreasing the value a little. I’m going to assume it’ll be in Los Angeles – as they’ve got the best chance. It was in Europe not long ago (2012) but it’s been a very long time since it was in the USA (1996, in Atlanta) and a USA Olympics is a huge money spinner for the IOC.
I would imagine that the cost of the 2024 Olympics rights would be at least $85m-$90m. Considering it’s not in our time zone, there’s a small hit that’ll mean the other bidders (Foxtel and 7 mostly – I don’t think Nine will take a punt on another Olympics) won’t want to pay too much. For the purposes of this hypothesis, let’s assume a $90m wire transfer to Lausanne gets the job done.
Buying the rights is only one portion of the expenses – you have to do all sorts of shit to actually broadcast it. That’s the next set of sums.
Putting on a broadcast on the scale of the Olympics isn’t gonna be cheap. The OBS handles all the front line work at the venue, so for our purposes, it’s all about getting every single second of every single event from every single camera at the Los Angeles 2024 Olympics to Australian eyeballs. The Olympics themselves actually does this already, via what they call the OVP – Olympic Video Player. It’s pretty much an app in a box that shows all the events. Taking the OVP footage and app adn re-skinning it is pretty much what they’ve done for 2016.
What I would want to do is take the OVP a step further and build our own platform, because the OVP sucks. The OBS produces the footage onsite, the OVP is the online distribution.
Streaming platform development – there’s various apps to make (Apple TV, iPad, iPhone, Android phone, Android tablet) a website and maybe even a couple of smart TV apps and PS4/Xbox One apps. Being the nerd I am, I reckon I could get a couple of mates together and make a better system than the drivel I’ve seen so far. Maybe a dozen or so developers and designers to work on it. Let’s say $2.5m in wages to get the platform made in 24 months, which I really don’t think would take 24 months to make.
Streaming platform operation – Someone has to orchestrate and maintain all the infrastructure. I reckon 3 people would cover it, so let’s over estimate and say 5. Probably $1m in wages here.
Bandwidth costs, very loosely estimated: 2 million people watch 40 hours of coverage in 5mbit/sec video (1080p H.265 – this is 2024 we’re talking about), that’s about 2PB of data transfer. On Amazon Cloudfront, on the Australian region, that’s $208,000. Again, very very rough estimate – other CDNs vary from about $50k up to $300k.
The streaming and product/promo website has to live somewhere, that’s just some basic AWS shit. $20k I reckon, way too much money, but I’m just guessing here.
There’d also be servers to transcode from the source feed, running something like Wowza on AWS infrastructure. Couple of EC2 instances shouldn’t cost more than $50k (the public hits the HLS/HDS streams via CDN) I reckon. There’d be heaps of instances though, as each sport/session would have it’s own stream and hence, it’s own EC2 instance pushing to the CDN. I think – I haven’t sat down and planned the actual streaming infrastructure.
On-site production – even though we’re just grabbing the OBS feeds, there’s still a lot of work required on-site to get the feeds from the venues to the broadcast centre, then to Australia, particularly with commentary. I don’t know how exactly (the OBS doesn’t publish this stuff online unforunately), but I guess there’d have to be a group of people sent over there, then some sort of method of getting the streams from the IBC to Australia. That could be satellite (I don’t think you’d be able to send every single venue’s output via satellite – not enough bandwdith), that could be a dedicated IP link (if it’s a 4K 60fps source for even venue, goddamn). I’m going to hazard a guess that all the ingestion and transcoding is done onsite, then sent to a CDN directly from the IBC. Would be cheaper and more reliable to do that, than send raw footage across the ocean to Australia and do it all here. This could be expensive – $4m sounds like what I think it would cost.
Commentary – the OBS doesn’t provide commentary, so you need to get experts in their field who can educate the viewers and tell the stories of the athletes. So you’ve got to send people there to watch it live, pay them to do it and pay them properly so they’ll do the background work required to be informed before they go.
Someone like 7 has heaps of production expertise for interviews, crosses, highlights packages etc. I don’t really care about any of that stuff. I literally just want the OBS feed, with commentary. There’s even the ability to have multiple audio streams with the video, so it’d be cool to licence say, ABC radio for the stuff they’re commentating. $1.5m here sounds like a good number, I really don’t know.
Marketing – you gotta tell people that the app exists! $5m in marketing staff, materials and placements, who knows what it really costs. Maybe the Olympics thing sells itself, I dunno.
Misc – you gotta have a few lawyers, accountants and a HR person or two involved as well I assume. Plus an office for them to work in as well as equipment and shit like that. Chuck $2m at this as a rough estimate.
Streaming platform development – $2.5m
Streaming platform operation – $1.5m
On-site production – $4m
Commentary – $1.5m
Marketing – $5m
Misc – $2m
All up there’s $16.5m in costs just to produce the content. I don’t know if this is even remotely accurate, but it’s a start. In total, including the rights ($90m), you’d be spending at least $106.5m to do this.
Now you need to make that money back, how many people have to watch this thing in order to at least break even?
An easy way to reduce costs is sell 7 the rights to FTA broadcast. Let them do whatever they want with it, we don’t care. If they’d pay $90m for full exclusive rights, including online, surely they’d pay $45m for TV only.
Plus I don’t think it would be too annoying if the app was sponsored. Like, say, McDonalds or Woolworths or Toyota or something. Some tasteful ads or whatever might get an extra $7m in our pockets. I totally reckon ads can be done right, without fucking up the stream and without ruining important events. There’s plenty of dwell time in all these streams, so that’s perfect for some ads. Maybe we also partner up with a telco to give away free unmetered access, like Optus, Vodafone or Telstra. Vodafone for example has ~5m customers, let’s say they’ll pay $1 per customer, that’s an extra $5m again.
So onselling the rights will get $45m, some branding/ads will get $7m and getting a telco onboard to give away to their customers, $5m. There’s $57m shaved off the $106.5m, leaving $49.5m left to make up via direct subscribers.
If we stick with the $20 rate as what 7 is charging now, that’s means we need 2,475,000 people to buy an account.
By 2024, Australia’s population will have grown. According to the ABS, Australia grows at around 1 person every 90 seconds. Australia’s current population is 24,151,757 – by August 2024, eight years away, that’s an extra 2,803,333 people (there’s 2.523e+8 seconds in 8 years, 2.523e+8 divded by 90 is 2,803,333), making Australia’s population around 27 million people.
If you reckon 10% of Australian will cough up cash to watch a fucking perfect implementation of a full suite of Olympics online streaming apps and website (if I made it, you bet your fucking arse it’ll be perfect and you’ll cry tears of joy at how beautiful it would be), that’s 2.7m accounts. That would bring in $54m in revenue, generating a modest $4m profit (you wouldn’t actually get $20 for each subscription, as there’s credit card/App store fees, but let’s ignore that for now), on which you’ll still have to pay tax anyways.
We don’t know the full stats of 7’s Olympics streaming, but I’d be surprised if 2.4m-2.7m take up premium access – that sounds like way too many people. According to Roy Morgan, in April 2015, there’s 1.89m Netflix accounts and 2.2m Foxtel accounts. You’d practically need every person who has Foxtel and/or Netflix to also want to pay for the Olympics.
If you up the price to $30 (totally worth it for the fucking awesome service your eyes will feast upon), that’s 1.65m accounts required. A bit more reasonable I reckon, but still, really really high.
If we tweak the other inputs a bit more – say Optus cough up $8m for premium access for their customers, sponsors kick in $10m and 7 reckon it’s worth $50m for them – for a total of $68m, leaving $38.5m to recoup. At $30 a subscription, that’s 1.29m subscribers required.
Well there you go, buy the rights for $90m, spend $16.5m to develop a platform and broadcast and market it, then resell the FTA rights to 7 for $50m, get $18m of sponsorship and manage to get about 1.5m Australians to pay $30 each. Easy!